From: Andreas Hartmann <andihartmann@01019freenet.de>
To: Larry Finger <Larry.Finger@lwfinger.net>
Cc: Justin Piszcz <jpiszcz@lucidpixels.com>,
linux-wireless@vger.kernel.org, linux-kernel@vger.kernel.org,
Alan Piszcz <ap@solarrain.com>
Subject: Re: Linux Wireless USB-Stick Question
Date: Thu, 04 Aug 2011 19:11:21 +0200 [thread overview]
Message-ID: <4E3AD2B9.6010905@01019freenet.de> (raw)
In-Reply-To: <4E39B82C.2010108@lwfinger.net>
Larry Finger schrieb:
> On 08/03/2011 02:53 PM, Justin Piszcz wrote:
>> Hi,
>>
>> Under Windows, you can achieve 10-15MiB/s..
>>
>> Under Linux, even with 150mbps USB wireless adapters, the max never
>> appears to
>> go above > 3-4MiB/s, to work around this, order more USB-wifi ticks
>> and run them
>> in parallel far away from each other with USB
>> extenders:
>>
>> box1:
>> -------------
>> wlan0 IEEE 802.11bg ESSID:"hidden"
>> Bit Rate=54 Mb/s Tx-Power=20 dBm
>> wlan1 IEEE 802.11bgn ESSID:"hidden"
>> Bit Rate=58.5 Mb/s Tx-Power=20 dBm
>> wlan2 IEEE 802.11bgn ESSID:"hidden"
>> Bit Rate=39 Mb/s Tx-Power=20 dBm
>>
>> box2:
>> -------------
>> wlan0 IEEE 802.11bgn ESSID:"hidden"
>> Bit Rate=58.5 Mb/s Tx-Power=20 dBm
>> wlan1 IEEE 802.11bgn ESSID:"hidden"
>> Bit Rate=52 Mb/s Tx-Power=20 dBm
>> wlan2 IEEE 802.11bgn ESSID:"hidden"
>> Bit Rate=52 Mb/s Tx-Power=20 dBm
>>
>> But I was curious if anyone had achieved > 10 MiB/s with any wireless
>> adapter
>> with Linux?
>>
>> Also, those native Linux USB adapters (carl) work good, so far.
>> With the patch provided earlier for the rt2800usb driver, it is no longer
>> crashing under 3.0 so I put two of them on a single box plus a carl
>> based one,
>> now I get better I/O, e.g. 4MiB/s x 6 = 24MiB/s.
>
> On a 150 Mbps connection running the following script
>
> #!/bin/sh
>
> dest="sonylap" # set the servername
>
> while true ; do
> netperf -t TCP_MAERTS -H $dest
> netperf -t TCP_STREAM -H $dest
> netperf -t TCP_SENDFILE -H $dest
> done
>
> I get the following for a D-Link DWA-130 containing a Realtek RTL8192SU
> with driver r8712u:
>
> finger@larrylap:~/bcm_git/vendor-driver/5.10.56.46> ~/netperf.sh
> TCP MAERTS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to sonylap
> (192.168.1.50) port 0 AF_INET
> Recv Send Send
> Socket Socket Message Elapsed
> Size Size Size Time Throughput
> bytes bytes bytes secs. 10^6bits/sec
>
> 87380 16384 16384 10.04 53.52
> TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to sonylap
> (192.168.1.50) port 0 AF_INET
> Recv Send Send
> Socket Socket Message Elapsed
> Size Size Size Time Throughput
> bytes bytes bytes secs. 10^6bits/sec
>
> 87380 16384 16384 10.03 55.58
> TCP SENDFILE TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to sonylap
> (192.168.1.50) port 0 AF_INET
> Recv Send Send
> Socket Socket Message Elapsed
> Size Size Size Time Throughput
> bytes bytes bytes secs. 10^6bits/sec
>
> 87380 16384 16384 10.06 65.26
>
> I claim that 50-65 Mbps is pretty good.
This means about half of fast ethernet (5-6,5 MB/s). It's too
insufficient for 802.11n / 40 MHz.
rt3572sta (Linksys WUSB600N v2) against rt2860 based AP (rt2800pci) -
from one room to another:
Do 4. Aug 18:48:27 CEST 2011
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to srv (1.1.1.1)
port 0 AF_INET
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
87380 16384 16384 10.07 82.51
TCP MAERTS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to srv (1.1.1.1)
port 0 AF_INET
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
87380 16384 16384 10.02 128.94
TCP SENDFILE TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to srv (1.1.1.1)
port 0 AF_INET
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
87380 16384 16384 10.09 77.47
rt2800usb doesn't work at all with WUSB600N v2.
ar9285 (ath9k) against rt2860 based AP (rt2800pci) - through reinforced
concrete floor:
Do 4. Aug 19:03:04 CEST 2011
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to srv (1.1.1.1)
port 0 AF_INET
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
87380 16384 16384 10.10 80.23
TCP MAERTS TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to srv (1.1.1.1)
port 0 AF_INET
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
87380 16384 16384 10.06 80.82
TCP SENDFILE TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to srv (1.1.1.1)
port 0 AF_INET
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
87380 16384 16384 10.10 78.51
-> there is room for optimization! I don't think, that the rt3572sta is
the best solution absolutely seen :-). Unfortunately, I don't have a
windows machine to compare ... . Maybe, I could get one to test ... .
Andreas
next prev parent reply other threads:[~2011-08-04 17:11 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-08-03 19:53 Linux Wireless USB-Stick Question Justin Piszcz
2011-08-03 21:05 ` Larry Finger
2011-08-04 17:11 ` Andreas Hartmann [this message]
2011-08-04 15:09 ` Stanislaw Gruszka
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4E3AD2B9.6010905@01019freenet.de \
--to=andihartmann@01019freenet.de \
--cc=Larry.Finger@lwfinger.net \
--cc=ap@solarrain.com \
--cc=jpiszcz@lucidpixels.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-wireless@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).